Development of Estimation Scheme for Soft Robotics System

Sensory data plays a significant role in perception and control of soft robots which provides the basis of informed decision-making. However, inherent softness in soft robots challenges the conventional ways of integrating sensors on them as soft robots have infinite degrees of freedom. An alternative approach is to obtain these sensory data through indirect sensing, which is to employ a filtering scheme that uses the robot’s dynamics and available sensor measurements to estimate variables that would have been measured by physical sensors. In this thesis, several modelling and estimation schemes are developed for multiple soft robot platforms.